Approximately 7,000,000 Americans suffer from stroke, resulting in significant morbidity and mortality. The majority of strokes are caused by the blockade of blood vessels (ischemic stroke). Currently, tissue plasminogen activator (tPA) is the only thrombolytic drug approved by FDA for the treatment of ischemic stroke. However, its use is severely limited due to the increased risk of cerebral bleeding. tPA is known to increase cerebrovascular permeability by promoting the opening of the blood brain barrier (BBB) in a mechanism that depends on the generation of active forms of PDGF-CC in the brain, which in turn activate the PDGF receptor-? (PDGFR?) on perivascular astrocytes (Su, et al, Nat Med, 2008). Emerging evidence suggests that tPA mediates the activation of latent PDGF-CC. However, tPA exhibits poor proteolytic activity without a cofactor and does not support efficient activation of latent PDGF-CC in solution. The mechanism that activates PDGF- CC in vivo under the setting of ischemic stroke is unknown. We discovered that microglia facilitate efficient activation of latent PDGF-CC in a mechanism that depends on tPA, integrin Mac-1 and the LDL receptor- related protein-1 (LRP1). Indeed, we found that the tPA-induced BBB opening in vivo depends not only on LRP1 but also on Mac-1, whereas the BBB opening induced by the cleaved/activated forms of PDGF-CC does not require either Mac-1 or LRP1. Based on these preliminary results and our published observation that Mac- 1 forms a complex with tPA and LRP1 (Cao, et al, EMBO J, 2006), we hypothesize that activation of latent PDGF-CC by microglia is dependent on the assembly of the tPA/Mac-1/LRP1 multi-protein complex. To test this hypothesis, we will perform structure-function analysis to define the molecular basis underlying Mac-1 interaction with tPA and LRP1. Guided by this structural information, we will design specific Mac-1 mutants that will not interact with tPA or LRP1, and tPA variants that possess full fibrinolytic activity but lack Mac-1 binding. We will also design specific antagonists to disrupt the formation of the tPA/Mac-1/LRP1 complex. The ability of these Mac-1 and tPA mutants to interrupt microglia-mediated PDGF-CC activation will be determined using a cell-based PDGF-CC activation system. In addition, the potential of the different antagonists to block PDGF- CC activation and thereby reduce BBB opening will be evaluated in a mouse model of experimental ischemic stroke. Completion of this project will generate new reagents that can protect the integrity of the BBB without compromising fibrinolytic activity of tPA. These reagents are also expected to specifically block the generation of active PDGF-CC without interfering with the normal functions of PDGFRs. As a result, these inhibitors should have fewer side effects than pan-PDGFR inhibitors, such as Gleevec, which is currently under evaluation in the I-STROKE clinical trial as an adjuvant therapy of tPA in human stroke patients. Thus, the information generated from this project could help us design better strategies to enhance the therapeutic efficacy of tPA while preserving the integrity of the BBB and therefore significantly impact clinical management of ischemic stroke.